Achieving radical change: a comparative study of public-sectoraccounting in Westminster and Scotland

Hyndman, N., & Liguori, M. (2018). Achieving radical change: a comparative study of public-sector accounting inWestminster and Scotland. Accounting, Auditing & Accountability Journal, 31(2), 428-455.https://doi.org/10.1108/AAAJ-04-2016-2527

Published in:Accounting, Auditing & Accountability Journal

Document Version:Peer reviewed version

Queen's University Belfast - Research Portal:Link to publication record in Queen's University Belfast Research Portal

Publisher rights© 2017 Emerald Publishing Limited.This work is made available online in accordance with the publisher’s policies. Please refer to any applicable terms of use of the publisher.

General rightsCopyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or othercopyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associatedwith these rights.

Take down policyThe Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made toensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in theResearch Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk.

Download date:14. May. 2020

1

Achieving radical change: a comparative study of public-sector accounting in

Westminster and Scotland

Authors’ Details

Noel Hyndman and Mariannuziata Liguori Queen’s University Management School Queen’s University Belfast Riddel Hall 185 Stranmillis Road Belfast BT9 5EE Northern Ireland, UK Tel: +44 (0)28 90974933 / +44 (0)28 9097 4491 E-mail: n.hyndman@qub.ac.uk/m.liguori@qub.ac.uk

Acknowledgements

The authors wish to express their gratitude to those individuals, whose comments and views are reported in this paper, who kindly agreed to be interviewed. The authors would also like to thank the Institute of Chartered Accountants in Scotland (ICAS) Foundation for their financial and research support of this project.

Author Biographies

Noel Hyndman is Professor of Management Accounting at Queen’s Management School, Belfast. A fellow of the Chartered Institute of Management Accountants, he was awarded a PhD by Queen’s University Belfast. His main area of research is in the field of public services, particularly accounting change, accountability and performance management in the non-profit and public sectors.

Mariannunziata Liguori is Senior Lecturer in Management Accounting and Director of the MSc Accounting and Finance at Queen’s Management School, Belfast. She was awarded a PhD, investigating processes of accounting change in the public sector, by Bocconi University, Milan, where she has previously worked. Her main area of research deals with processes of accounting change and with control practices in public-sector outsourcing.

2

Achieving radical change: a comparative study of public-sector accounting in

Westminster and Scotland

Abstract

Purpose – The paper focuses on strategies and ‘spoken discourses’ used to construct legitimation around change at the individual level. Comparing changes in financial accounting, budgeting and performance management at two government levels (Westminster and Scotland), it explores the use of legitimation strategies in the implementation of accounting change and its perceived outcomes.

Design/methodology/approach – Drawing on semi-structured interviews, six legitimation/delegitimation strategies are used to code the transcribed data. Patterns with the perceived outcomes of change are explored.

Findings – Changes introduced to enhance ‘rational’ decision making are often received as pushed by some source of authority. Regardless of the interviewees’ background and level, the results suggest that for radical accounting change to embed, it is necessary for it to be perceived as rational, rather than merely driven by authorisation-based pressures. Conversely, incremental change is associated with modest legitimation via rationalisation and delegitimation based on pathos and rationalisation.

Research limitations/implications – The study deals with actors’ legitimation strategies and perceptions of change. These may not correspond to actual substantial change. Taken-for-granted ideas often remain ‘under the radar’, therefore care must be taken in interpreting the results. The focus of the empirical study is on the UK, therefore conclusions are restricted to this context.

Originality/value – Existing studies struggle to explain organisations’ heterogeneity and practice variation; this study sheds light on how individual legitimation, which may lead to different organisational results, occurs. Differences in how actors interpret changes may be based on their position (central vs. devolved administration) and on their ownership of the changes.

Keywords – accounting change, public sector, legitimation strategies.

Article classification – Research paper

3

Achieving radical change: a comparative study of public-sector accounting in Westminster and Scotland

Introduction

Change can be imposed, but if it is not seen as legitimate (or having substantive justification) in the context in which it is applied, it is likely to be resisted. Previous literature, both in accounting and the public sector, has mainly looked at accounting change, focusing on organisation-wide factors (Kitchener, 1999; Burns and Scapens, 2000; Liguori and Steccolini, 2012; Liguori, 2012b). There is limited work on how accounting change is legitimated at the individual level, within and across organisations, and how legitimation strategies affect the reproduction and outcome (or significance) of change; a few studies only pointing to the importance of legitimation rather than addressing it directly (Broadbent, 1992; Covaleski et al., 2003). Comparing the changes in financial accounting, budgeting and performance management (collectively referred to as ‘accounting’ for convenience) that took place over the past 20 years at two different government levels (the central administration of Westminster and the Scottish devolved administration), the paper aims to contribute to this body of literature by exploring how organisational actors justify change, and whether their arguments are associated with the outcome of change. The study focuses on how legitimation strategies are used to construct a sense of legitimation around change at the individual level (Van Leeuwen and Wodak, 1999; Vaara et al., 2006). The paper sheds light on the translation and legitimation cues that may lead to different types of organisational change by exploring patterns and relationships between legitimation strategies used to interpret the changes and the perceived outcome (be it ‘radical’, or merely ‘incremental’) of change. Moreover, it investigates change and legitimation dynamics not only across different departments (central finance vs. operational departments), but also at different government levels (Westminster and Scotland). Differences emerge in relation to the level of ownership that the organisational actors perceive themselves as having over the changes being implemented.

The paper is organised as follows: the next section presents a review of the relevant literature on legitimation strategies and outcomes of change; this is followed by the methodology, and a summary of background and context of the UK accounting systems. The subsequent sections present the results for Westminster and Scotland. The penultimate section discusses the main findings, with the final section drawing the conclusions and proposing avenues for further research in the area.

Legitimation strategies and accounting

Both accounting and public administration scholars have increasingly recognised that global accounting reform trends have been translated with significant local variations (Barzelay, 2001; Pollitt and Bouckaert, 2011). It has been suggested that changes do not always yield the expected results (Hood and Peters, 2004; Lapsley, 2009) and they tend to produce a layering, rather than a replacement, of practices over time (Hyndman et al., 2014). Much attention has been devoted to considering the dynamics and processes of accounting change at the organisational level (Broadbent, 1992; Burns and Scapens, 2000; Connolly and Hyndman, 2011; Liguori and Steccolini, 2012); less has been done to consider the role that actors’ understanding of change can play at the individual level. How accounting changes are perceived, translated and legitimated at the micro level, and how this affects the significance of change, has received limited attention. In order to explain why the introduction of a new practice succeeds or fails in a certain institutional field, however, it is essential to understand how individual actors legitimate or delegitimate change and the arguments they mobilise for it. The sources of change and legitimation are made available in the field by the institutional players themselves, who engage in a process that, in institutional theory, has been alternatively defined as ‘theorisation’ (Strang and Meyer, 1994), ‘translation’

4

(Czarniawska and Joerges, 1996), ‘editing’ (Sahlin-Andersson, 1996) or ‘legitimation’ (Green and Li, 2011).

The legitimation of change, in the context in which it is applied, is essential if change is to be accepted and resistance avoided. According to Scott (1995, p.45), legitimation ‘is not a commodity to be possessed or exchanged but a condition reflecting cultural alignment, normative support or consonance with relevant rules or laws.’ The specific context conditions define ‘which ideas are considered ‘sensible’, which constructions of reality are seen as ‘realistic’, and which claims are held as ‘legitimate’ within a certain policy at a specific time’ (Koopmans and Statham, 1999, p.228). In legitimating change, language plays an important role for its diffusion and implementation (Green, 2004). The rhetorical arguments and strategies set in place during change can strengthen or weaken the adoption of new practices and the way actors understand and rationalise the usefulness of the change itself (Hirsch, 1986; Green, 2004). Legitimation strategies are here defined as the discursive devices ‘used to (re)construct senses of legitimacy/illegitimacy’ (Vaara et al., 2006, p.789), i.e. micro-level discursive practices that are used or talked about when making sense of, and giving sense to, one or more changes. The ambiguity regarding the process and the outcome of change induces organisational actors to use rhetorical arguments to construct reality and gain legitimacy on certain organisational structures and beliefs (Creed et al., 2002; Suddaby and Greenwood, 2005). The more persuasive the supporting reasons justifying the introduction of a new practice, the more rational its adoption will be perceived, and ultimately accepted. However, how rhetoric and language shape the introduction and reproduction of change within organisations has been little investigated and, in particular, how legitimation (or delegitimation) strategies are used at the individual level (by individual actors) to achieve change is still unclear (Vaara et al., 2006; Green et al., 2008). More field-based micro-analysis has been advocated to reach a better understanding of the complexity and the contradictions that can arise in legitimating change (Vaara et al., 2006; Green et al., 2008). This can also help explain why the implementation of similar ideas and practices has often resulted in notably different results at the organisational level (Etherington and Richardson, 1994; Liguori and Steccolini, 2012).

Organisation study research has looked at the rhetorical legitimation of change, mainly focusing on particular fields, such as industrial restructuring and mergers and acquisitions (Vaara, 2002; Comtois et al., 2004; Vaara et al., 2006). Legitimation strategies have been found to be important: in processes of institutional change (Green, 2004; Suddaby and Greenwood, 2005); in actors’ struggles over controversial decisions, such as organisational shutdowns (Erkama and Vaara, 2010); in contested issues, such as climate change (Lefsrud and Meyer) and international corporate investments (Joutsenvirta and Vaara, 2015); and in policy-making and institutional change (Brown et al., 2012). In relation to accounting firms, Covaleski et al. (2003) studied the ‘Big Five’ to investigate how competing factions re-institutionalise societal expectations of proper professional behaviour to legitimate a transformation of jurisdictions. They found that accountants tend to use abstract rhetorical arguments to legitimate and justify changing jurisdictions.

Other accounting-focused studies have mainly looked at the actors’ attribution of meanings, limiting the analysis to the cognitive attributes of accounting narratives and documents (see Aerts, 2001; Tsang, 2002; Clatworthy and Jones, 2003). Using a theoretical framework different from the one adopted in this paper, some have also utilised legitimacy theory to investigate environmental disclosure as a mandatory practice or as a social contract (Mobus, 2005). For instance, Cho (2009) examined the environmental disclosure decisions and practices of one of the largest oil and gas companies in the world. He concluded that communication played an essential role to legitimate the company’s actions, and social and environmental disclosures remained a powerful legitimacy device for greater external accountability. Hrasky (2012) explored whether Australian companies adjusted their footprint-related disclosure responses; she concluded that, while carbon-intensive sectors were pursuing a moral legitimation strategy, the less carbon-intensive sectors relied more on symbolic disclosure. Buhr (2002), on the basis of analysing environmental reporting in two

5

Canadian companies, investigated the moral and pragmatic strategies used to justify environmental issues and sustainability reporting. These studies have largely focused on the external legitimation of change, rather than intra-organisational, individual, legitimation. Moreover, previous research has mainly dealt with moral and normative aspects of legitimation, often from the external stakeholders’ point of view. We claim there is still a lack of comprehensive studies that examine the broad spectrum of rhetorical and legitimation arguments that organisational actors have available (and may draw upon) to make sense and justify accounting change. This study aims to fill this gap through the use of a different theoretical lens (the one of legitimation strategies), spotlighting individual legitimation strategies; an approach rarely adopted when investigating accounting phenomena.

Change and legitimation strategies

Despite the contribution of the interdisciplinary and critical movement (Neu et al., 2001), a wide spectrum of society often sees accounting as a neutral technology that, in the hands of rational decision makers, can be deployed to generate and draw attention to information to support decision makers in making better decisions (Knights and Collinson, 1987). However, the implementation of accounting changes may be framed in a variety of ways, ranging from value-neutral procedures to ideologically-charged instruments. Change may be contested, resisted or transformed by organisational groups (Motion and Leitch, 2009). The implementation of accounting techniques is not a mechanistic, simplified introduction of a new practice into formal organisational structures, procedures or mechanisms. Rather it represents a translation of general ideas adapted to local circumstances (Brunsson, 1989). To investigate the role language and rhetorical arguments play, different research approaches have often been brought together under the label ‘rhetorical institutionalism’ (Green and Li 2011) and mainly include research on framing (Fiss and Zajac 2006), discourses (Phillips et al. 2004) and rhetoric (Green 2004) within institutional fields. This study finds its theoretical roots in this latter group.

When dealing with legitimation and rhetorical techniques and the introduction of new practices which potentially replace, or undermine, existing structures, Green (2004) and Green and Li (2011) propose the identification of a source of ‘endogenous change’ based on the use of different legitimation strategies. It should be noted that this is different from the legitimacy theory approach, which focuses on the process of legitimation through which an organisation seeks approval (or avoidance of sanction) from society, frequently through voluntary external disclosures (Suchman, 1995). In contrast to this, focusing on legitimation strategies stresses the importance of individuals and their construction of reality during change. This is mirrored in their rhetorical arguments accompanying the introduction of change and affecting the way it is interpreted. Following Vaara et al. (2006) and Van Leeuwen and Wodak (1999), new techniques and ideas can be put forth relying on one of five discursive strategies (or a combination of these): authorisation, rationalisation, normalisation, moralisation or narrativisation. Authorisation strategies refer to legitimation through reference to authority (such as government or politicians), custom, law and persons upon whom institutional authority of some kind has been bestowed. Rationalisation is related to legitimation by assessing the benefits that a certain course of action can bring. In this study, for instance, it refers to issues such as rational decision-making practices, cost-benefit analysis, effective planning, the importance of resources and skills, etc. Normalisation legitimates by exemplarity that can involve past or current cases that are deemed as something ‘normal’ and thus used as standards. In this study with respect to the use of normalisation strategies, reference was frequently made to professions and professional standards (which are particularly important in areas such as accounting). The use of such strategies is likely to be affected by the actors’ background and education. Previous studies have highlighted the importance of the accounting profession and background in the design and implementation of reforms (Humphrey 2005;

6

Christensen and Parker 2010). Indeed, accountants can be viewed as privileged ‘communities’ able to understand the newly introduced technologies, ‘insiders’ who have knowledge of technical accounting issues and tend to be supportive of ‘modern’ technologies of accounting (Ezzamel et al., 2005). Moralisation refers to legitimation by reference to specific value systems and what is considered ethical in a specific field. In an accounting context, this is not only linked to the purpose of serving external and internal stakeholders’ information needs, but also to the identification and judgement of accounting scandals or cases of professional misconduct. Finally, narrativisation is about legitimation conveyed through narratives; telling a story, indeed, can provide evidence of acceptable, appropriate or preferential behaviour for current changes and future practice. More specific examples of these strategies in the context under study are provided in the methodology section.

Authorisation, rationalisation and normalisation strategies align with Green’s (2004) notion of logos (legitimation of change through the adoption of rational arguments from different sources), while moralisation strategies align with what Green (2004) calls ethos (legitimation through credibility, moral authority or tradition). He also identifies a route of legitimation via pathos, i.e. legitimation by appealing to emotions (Green et al., 2008; Green and Li, 2011), which does not necessarily coincide with the above strategies. The variation in the acceptance of a particular change may be influenced by the deployment of specific arguments in particular situations. However, legitimation strategies are not always intentional or conscious, and their use usually decreases with time, when justifications are less needed as a change becomes accepted or a new practice is taken for granted (Green, 2004).

Legitimation strategies influence the organisational conditions that help restore or justify social status quos, or even destroy them through delegitimation (Van Leeuwen and Wodak, 1999). In order to undermine or stop change, different actors might use antagonistic framings or destructive strategies, criticising the validity or ridiculing the source of informationi (Van Leeuwen and Wodak, 1999; Lefsrud and Meyer, 2012). Green et al. (2008) opine that different actors tend to use the same rhetorical sequences during successive periods of change. Moreover, movements from the status quo, or stable institutional field arrangements, are often initially brought about by emotional appeals to redirect limited social attention. Pathos arguments may have a part to play, particularly at the beginning of the institutionalisation process, because institutions have an inherent social inertia. Legitimation is frequently viewed as a prerequisite for the institutionalisation of new ideas and practices (Kitchener, 1999). At the same time, institutionalisation contributes to legitimation because, over time, established and widely spread ideas and practices no longer require the use of specific legitimation strategies (Green, 2004; Green and Li, 2011).

Accounting change and its outcome

The way organisational actors perceive, talk about and legitimate change is inevitably interwoven with the way they react to and implement it. We posit that the legitimation strategies used to talk about and understand change will affect the outcome of the process of change, i.e. the way change is embedded and its significance within an organisation. This relationship has never been investigated before. However, the different embedding often achieved even in the presence of similar change requirements cannot be explained by the arational mimicry suggested by the traditional strands of institutional theory, which has been criticised for paying scant attention to individual micro processes (Lounsbury, 2001). Within the same institutional realm, archetype theory provides a more comprehensive framework able to address such limitations also in the study of accounting. Archetype theory (Hinings and Greenwood, 1988; Greenwood and Hinings, 1996) represents a powerful tool to fine-grain the level of analysis, and identify and interpret both the outcome and pace of change (Liguori and Steccolini, 2012; Liguori, 2012a and 2012b). It allows distinguishing between different outcomes (radical vs. incremental) and different levels (structures

7

and systems vs. ideas and values) of change, adopting a micro-level perspective. While a number of accounting (and other) studies have adopted these lenses to study change (Kitchener, 1999; McNulty and Ferlie, 2004; Liguori and Steccolini, 2012; Liguori, 2012a and 2012b), they neglected (or downplayed) the role of individual rhetoric and discursive elements, stopping at the discussion of organisational factors and their interaction. The present study attempts to shed more light on these less investigated elements.

An archetype is defined as a set of structures and systems that reflects a certain interpretive scheme (Greenwood and Hinings, 1993). Organisational structures and systems, such as management techniques, performance dashboards or financial accounting systems, can be seen as embodiments of ideas, beliefs and values that underlie their existence and functioning. These very ideas constitute what archetype theory defines as an ‘overarching and prevailing interpretive scheme’. An archetype is, thus, a set of ideas, belief and values that shape prevailing conceptions of what an organisation should be doing, how it should be doing it and how it should be judged, combined with structures and processes that implement and reinforce those ideas. Change is not indiscriminate, but happens differently on the basis of the levels involved (i.e. systems and structures and/or interpretive schemes). The consistency of change with the existing archetype defines the distinction between incremental and radical outcomes of changeii. The former takes place when organisations modify their structures and systems in a way that is consistent with the pre-existing archetype, i.e. structures and systems change, but the old interpretive schemes are used to make sense of them. Radical change, instead, involves a shift and adjustment in both structures (and systems) and interpretive schemes, which both move from a pre-existing archetype to a new one (Greenwood and Hinings, 1996).iii

Previous studies adopting an archetype perspective showed that, although confronted with similar environmental pressures, organisations often display different patterns of change (McNulty and Ferlie, 2004; Hammerschmid and Meyer, 2005). The movement identified in the public sector, in particular, is from a bureaucratic to a managerial archetype (for details of these archetypes in accounting see Liguori and Steccolini, 2012; Liguori, 2012a and 2012b). This literature, however, stopped at the consideration of organisational factors, such as skills, leadership and resource distribution, and failed to include individual features of translation and legitimation of change. Siti-Nabiha and Scapens (2005) acknowledged the overlapping and consistency between the Burns and Scapens’ (2000) model (often used in accounting research) and Greenwood and Hinings’ (1996) framework. More recently, Ter Bogt and Scapens (2014) have used the notion of archetype, as a form of coherent structures, ideas and practices to investigate the relationship between institutions and actions. Their study, however, focused on the level of field-wide institutional logics, rather than on the actors’ translations during change. This paper focuses on this latter aspect by analysing the relationship between actors’ legitimation strategies (used to understand change) and the outcome of change actors themselves perceive and contribute to shape.

Methodology

This research investigates how organisational actors justify change, and how their arguments are associated with the perceived outcome of change. It is based on the comparative case studies (Pettigrew, 1990; Ragin, 2000) of the Westminster central government and the Scottish devolved administration. International comparisons of the public-sector managerial reforms that took place from the 1990s onwards indicate that they have been implemented at a quicker pace and more enthusiastically in some countries compared to others, with the UK viewed as a high-intensity, early adopter (Hood, 1995; Pollitt and Bouckaert, 2011). Managerial reforms deeply involved accounting systems and also brought about a process of decentralisation and devolution of government, which has fostered the increased independence of ‘devolved administrations’. The

8

UK, in particular, through the comparison of Westminster and Scotland, provides an interesting case to investigate how similar accounting changes can be implemented and legitimated differently at different organisational and government levels, even within the same state.

The paper explores the changes in financial accounting, budgeting and performance management in the public sector that started in the 1990s, the reactions to these and the legitimation strategies used to describe them at two organisational levels: the finance departments (for simplicity identified as Westminster Finance - WF - for Westminster and Scottish Finance – SF - for Scotland), who firstly pushed the changes; and the operationally-focused departments, largely concerned with the delivery of services. Financial accounting, budgeting and performance management are at the core of the managerial reforms that were introduced in the public sector over the past few decades. Changes in these three areas have often been implemented in an integrated fashion and together (for a detailed discussion, see Hyndman and Lapsley, 2016). In both Westminster and Scotland, two operational departments (identified as WD for Westminster and SD for Scotland) were considered: rural development and higher education. While the finance departments had the task of translating the political ideas into technical requirements and pushing the accounting changes down into the rest of the organisation; the more operationally-focused departments had to translate the new changes and legitimate (or delegitimate) their use in everyday activities. As the implementation paths of the same changes may vary within, and be perceived as more applicable to, differing activities of government, the two line departments were selected following a maximum variation sampling (Patton, 2002), on the basis of their different output measurability. Drawing on the studies of Brown and Potoski (2003) and Cristofoli et al. (2010), activities that are more prone to commercial valuation, such as environmental and agricultural ones (see, for instance, insect control, snow ploughing, etc.) are expected to show higher output measurability than those involving care of people or development of skills, such as cultural and arts programmes (and, we posit, education services).

In Westminster, the main originator of change was HM Treasury (WF), one of 24 ministerial departments within the UK central government; it was also responsible for developing and executing the UK government’s public finance and economic policy. A range of changes have been introduced under the auspices of WF in order to facilitate effective management in the more-operationally-focused departments. On the side of the receivers of change, Westminster rural development falls within the competencies of the Department for Environment, Food & Rural Affairs; this is responsible for general policy and regulations on environmental, food and rural issues. Westminster higher education is under the Department for Business, Innovation & Skills. This department engages with universities and colleges to facilitate them in providing high quality teaching and research, with funding in research informed by a Research Excellence Framework (REF: a performance measurement framework that utilises expert review to assess, largely, research outputs and impacts). In Scotland, since 2010, the government directorates are responsible for progressing their strategic objectives. The finance functions of the government (referred to as SF) currently fall under the auspices of the Cabinet Secretary for Finance, Employment and Sustainable Growth. Rural development in Scotland is the responsibility of the Agriculture, Food and Rural Communities Directorate, with higher education falling under the Employability, Skills and Lifelong Learning Directorate. The 19 autonomous colleges and universities are all funded by the Scottish government via the Scottish Funding Council, established in 2005.

Exploring legitimation strategies and change required asking organisational actors about their experiences and perceptions. This involved conducting semi-structured interviews to reconstruct the understanding of the key actors as to the main accounting changes over the last two decades, and the related (implicit) rhetorical strategies of legitimation or delegitimation. Questions were asked about the most relevant accounting changes (and sources of such changes) they had to face, as well as people’s reactions and views of such changes, together with their

9

perceived impacts. The interviews were conducted with 18 senior managers in Westminster and 16 in Scotland, who were responsible for both policy and accounting issues. The interviewees are identified later in any quotations by their designations: Westminster Finance from WF1 to WF4, Scottish Finance from SF1 to SF8, WD1 to WD14 and SD1 to SD8 for the operational departments’ interviewees in Westminster and Scotland respectively. The Scottish interviewees had significant knowledge and experience of accounting requirements and changes introduced both before and after the 1999 devolution (see later for explanation regarding devolution).

The interviewees, who were selected to get information-rich cases from those directly involved in the process of change, were identified through snowball sampling (Patton, 2002). In terms of professional background, in Westminster all the WF interviewees had an accounting background and professional qualification, while 10 out of 14 had a similar background in WD. Six out of eight of the SF interviewees had an accounting background and related professional qualification, as did five out of eight in SDiv. Because of their extensive service and long institutional memory (on average, 15 years of tenure in Westminster and 20 years in Scotland), the majority of the interviewees were able to reflect on the changes over a considerable period of time (covering either all of the period under analysis, or a vast part of it). As the interviewees’ reconstruction of events could have been influenced and limited by their memory and ability to recollect changes that took place in the past, we relied on triangulation of investigators, informants and data sources (Patton, 2002). In the short term sources of change may appear blurred, however, in the long term, relationships may become clearer to organisational members. As found in other studies (Liguori and Steccolini, 2012), during the research, several of the interviewees admitted that the interview was also useful to them in providing a clearer understanding of the actual meaning of a change that had been implemented. This strengthens the validity and credibility of our findings. The process of ex-post rationalisation allowed both researchers and interviewees to identify the real epoch-making changes, which may have been less visible at the time they were being implemented. Similarly, a more comprehensive evaluation of the changes became possible. It is important to note that this study shares the view of processes of change as continuous (Poole et al., 2000): adjustment and changes to some of the accounting systems were present and being discussed also at the time the interviews were carried out (see, for instance, the Clear Line of Sight – CLOS – project, Table 1).

The interviews were carried out between the end of 2012 and spring 2014, and each lasted, on average, one hour. Each was recorded and transcribed in full for coding. A preliminary coding scheme was developed based on the relevant literature on change and legitimation strategies. Following an iterative process, the coding scheme was refined and applied to ensure data consistency. The coding scheme was applied to each of the interviews independently by the two researchers, with all cases of disagreement being reviewed and resolved as a team. Data coding and analysis were supported by ATLAS.ti 6. A specific argument was coded when a legitimation strategy was used with regard to one of the possible areas of accounting change (financial accounting, budgeting or performance management) or relating to the reforms as a whole. A question about the main changes perceived by the interviewees was used as ‘ice breaker’, followed by prompts for each of the different possible areas of change; this was to ensure that all the changes identified in the literature were covered/discussed. The range of implemented changes and their classification refer to those reconstructed through official documents, interviews and literature review (see Table 1). Official documents (identified utilising a variety of contacts, including high-level HM Treasury and Scottish Finance networks) included, for instance, the Green and White Papers and the CLOS report for Westminster, as well as the National Performance Framework (NPF) scheme for Scotland.

In distinguishing the legitimation strategies, the typologies proposed by a number of writers were combined and resulted in six legitimation/delegitimation strategies being operationalised: authorisation (references to political or mimetic pressures, financial crisis, fiscal requirements, external stakeholders, market pressures, the role of the EU, Government, law and

10

regulation, internal management, etc.); rationalisation (such as references to managerial and accounting reforms, the importance of culture, effective planning and decision making, skills and education, resources, IT services, etc.); normalisation (based on references to profession, comparison between public and private sector, etc.); moralisation (such as transparency, gender equality, social and environmental sustainability, good administration, etc.); narrativisation (through the use of organisational ‘stories’ or history, accounting scandals or exemplars of behaviour); and pathos (references to the role of personal or organisational commitment to change, career dedication, patriotism, etc.). Legitimation strategies were coded as ‘1’, whereas delegitimation strategies were coded as ‘2’. As highlighted in previous studies, frequently, a number of strategies were combined in the same argument by the interviewees to help them make their point, or strengthen their particular view on a change. The number of occurrences for each strategy was computed so that a repetition of the same argument within the same answer was only counted once. The absolute counts and relative prevalence of each legitimation strategy by departmental level and government is presented in Table 2. Since within the same argument, a number of strategies could be used, the number of strategies counted in Table 2 is different from the total in Table 3 (where only one outcome of change per argument was coded). The same is true for Table 4, where more strategies were coded in each argument and for each outcome of change identified.

Finally, in identifying the perceived outcome of change, the research drew on the classifications proposed by Greenwood and Hinings (1996), Liguori and Steccolini (2012) and Liguori (2012a). Questions were asked to enquire indirectly about the actual use of the new accounting tools, the decision-making criteria used and the agreement with the new managerial ideas. Following Liguori and Steccolini (2012), radical change was identified whenever it was possible to highlight a consistent movement from the previous bureaucratic archetype to the new managerial one in both accounting structures and related interpretive schemes (i.e. the related ideas, values and taken-for-granted actions, as understood by the interviewees). A change only at the level of accounting structures and systems was defined as incremental (for instance, in the case new RAB systems or performance measures were formally introduced, but these were not actually used, agreed upon or understood by those using them). The lack of change in the interviewees’ ideas and interpretive schemes, accompanied by fairly minor (if any) changes in the organisation’s accounting systems was coded as ‘no change’v (Liguori, 2012b). The analysis of patterns of co-occurrence between the perceived outcomes (radical, incremental or no change) and areas (financial accounting, budgeting, performance management or reform as a whole) of change is presented in Table 3, while Tables 4 and 5 present emerging patterns between outcomes of change and legitimation strategies. The Appendix provides examples of how the coding scheme was applied to the interview transcripts to identify different strategies and change outcomes.

Accounting changes in the UK: setting the context

The UK consists of four separate countries (England, Northern Ireland, Scotland and Wales) within one main political unit. Traditionally, government within the UK has been founded on concepts such as a unitary state and supremacy of a parliament at Westminster. These ideas have been challenged by the process of devolution, this being defined as the transfer to a subordinate elected body, on a geographical basis, of functions previously exercised by Westminster ministers and the Westminster parliament (Bogdanor, 2001). Since 1999 there have been devolved administrations, in Northern Ireland, Scotland and Wales, with these remaining part of the UK and Westminster retaining ultimate power. The areas in which the Scottish parliament has devolved powers (legislative authority) include health, education, local government, justice and agriculture. The UK parliament’s powers, with regards to Scotland, include defence, foreign policy, financial and economic matters, energy and medical ethics (Scottish Parliament, 2016).

11

The UK is a common law country and the vast majority of financial accounting, budgeting and performance management changes do not pass through law, but rather are implemented via administrative acts (for more detail, see Hyndman et al., 2014). This is not to say that specific accounting reforms are announced and then implemented without adjustment and in accordance with initial timetables. Indeed, often a degree of evolution and shape adjustment takes place between initial announcement and the final decided changes. As a result of the managerial reforms that took place in the 1990s, accounting systems in all parts of UK central government changed considerably. Initially, the UK central government featured: cash budgeting with strict annuality requirements, cash accounting, and limited performance measurement and performance management. Since then it has progressively moved significantly along a continuum of reform in financial accounting, budgeting and performance management (Table 1). The earliest major changes were seen with respect to performance management. The Financial Management Initiative called for managers in central government to have ‘a clear view of their objectives, and means to assess, and wherever possible measure, outputs or performance in relation to those objectives’ (HM Government, 1982, p.5). This high-minded aspiration was a recurring theme of the Next Steps Initiative, the Citizen’s Charter and developments in education and health care in the 1980s. Performance management and reporting were also integral parts of the development of Resource Accounting and Budgeting (RAB) changes in the mid-1990s (see below) and came again to the fore in the introduction of Public Service Agreements (PSAs) and Service Delivery Agreements (SDAs) in 1998. Central government departments were held accountable for service delivery through the targets set out in their PSAs and SDAs. In 2004, SDAs were replaced by performance-focused Delivery Plans for each department.

As can be seen in Table 1, since 1980 financial accounting has moved from its traditional cash base to a more commercial accruals (or, in the language of RAB, ‘resource’) base. The arguments for using the accruals model, in place of cash accounting, gained prominence and a move from the cash basis was agreed for central government (see the Green Paper – HM Treasury, 1994, followed by the White Paper – HM Treasury, 1995). This was implemented under the title of ‘RAB’. In addition, on the management accounting side, a move to resource budgeting, in order to make the management accounts align with the external accounts, was made. The position of accruals accounting was further embedded by the decision in 1998 to produce Whole of Government Accounts (WGA) (a consolidated set of accruals financial statements for the UK public sector, consolidating around 1,500 bodies), with the first set of WGA being produced in 2011. In addition, in 2007, it was announced that the financial statements of public sector organisations would be prepared using International Financial Reporting Standards (IFRS); this change was implemented by 2009.

In the realm of budgeting, major changes have also occurred. As mentioned above (and shown in Table 1), the move from cash budgeting to resource (accruals) budgeting was announced in the mid-1990s (‘live’ by 2003). In addition, annuality (the requirement of budget allocations to be spent by the financial year-end or be surrendered) was abolished in 1997 at the central government departmental level, with end-year flexibility (EYF) being permitted (allowing the carry forward of unspent resources from one year to the next). However, as a consequence of financial pressures, this flexibility was removed in 2011. To provide greater long-term focus, in 1998, Treasury-led spending reviews were initiated to set firm and fixed spending budgets over several years for each government department rather than rely on single-period budgeting.

Insert Table 1 here

Variations in Scotland

12

Up until the late 1990s the UK operated as a very centralised state, with substantial powers exercised by the Westminster parliament in London. However, since 1999, Scotland has had a devolved parliament with full legislative competence, that is to say it holds primary legislative powers and potentially the power to legislate in any non-reserved area of policy. In preparing for devolution, a series of consultation working parties was organised by the then Secretary of State for Scotland. One of these groups was the Financial Issues Advisory Group (FIAG), which was given the remit of reporting on how best the new Scottish parliament could exercise budgetary control. While the majority of UK financial accounting rules and changes also applied to Scotland, the FIAG observed that it would not be appropriate for the Scottish parliament to merely adopt the existing practices of budgetary scrutiny of the UK parliament at Westminster (Scottish Office, 1998). It highlighted that the main objective of the budgetary system of the new Scottish parliament should be related to transparency. The mechanism to achieve this, advocated by FIAG and adopted by the Scottish parliament, was a three stage process through: scrutiny of strategic priorities for expenditure; examination of the draft budget; and approval of the expenditure plans in the budget documents.

In terms of performance management, Scotland experienced similar influences as the rest of the UK prior to devolution. Subsequently, while the focus on such has been at least as intense as in the rest of the UK, the manner in which it has been applied has been different. Perhaps the most notable outworking of this was seen in the Scottish spending review of 2007, which introduced a NPF. The NPF, inspired by a model developed by the state of Virginia in the USA, aims to provide a single framework with a ten-year vision and based on delivering outcomes, rather than inputs and outputs that improve the quality of life for people in Scotland (Scottish Government, 2011).

Finally, despite Scotland having responsibility for the allocation of their budget and a distinctive performance management approach, overall financial arrangements for accounting and reporting need ‘to be consistent with the Whitehall regime for planning and controlling public expenditure’ (Lapsley and Midwinter, 2008, p.78). Thus, when Westminster introduced RAB, the Scottish parliament followed suit. Similarly, Westminster’s decision to introduce EYF in 1997 was also applicable to Scotland. Similarly, the abolition of EYF and its replacement with the Budget Exchange system was applicable to Scotland (HM Treasury, 2011). There has been a large degree of convergence on the main financial accounting and budgeting processes used by both the Westminster and Scottish parliaments. Where divergences have occurred, these have mainly related to the more fine-grained application of devolved responsibilities, rather than the main tools of planning and control, and their related accounting instruments, utilised within the UK as a whole.

Legitimation of accounting changes in Westminster

Overall, the majority of Westminster interviewees were accountants, often with private sector experience (two out of four interviewees in WF; five out of 14 in WD) before joining central government. In WF, the interviewees mainly referred to strictly financial accounting changes (such as the introduction of accruals accounting or WGA) during their interviews, with much lower emphasis on budgeting issues and, especially, on performance management. When discussing the different areas of accounting change, WF concentrated strongly on legitimation rather than delegitimation strategies. As can be seen in Table 2 (which shows numbers and percentages of legitimation/delegitimation strategies per organisational level and government, Westminster and Scotland shown separately), delegitimation strategies (AUT2, NAR2, NOR2, PAT2 and RAT2) were only mentioned in 47 cases out of 315 counts (about 15% of all the cases). In other words, changes were largely interpreted through positive statements and with limited opposition. In terms of legitimation, the most-frequently employed strategy concerned rationalisation (RAT1 24.4%,

13

Table 2), followed by authorisation (AUT1 22.2%). With respect to rationalisation, the interviewees reflected on the need for the new forms of account to support upward communication (to politicians) and within-department communication (to managers) as a basis for discussion and better decisions, in particular when talking about RAB. This strategy, especially in the form of effective planning, also represents the most used in Westminster generally (28.4%, Table 2). Interestingly, delegitimating strategies linked to rationality (RAT2) accounted for only 5.7% of the cases in WF. This was not particularly with respect to the unsuitability of specific individual tools, but largely in relation to implementation problems connected to the initial misalignment between financial accounting and budgeting systems.

In the case of authorisation, the main source of authority referred to was political or government. Authority was seen to be pushing or supporting the changes, and, in most cases, seemed to be appreciated (perhaps viewed as providing welcomed back-up to accounting changes that were considered ‘rational’ or ‘normal’ by the WF interviewees). One interviewee, who had lived through the change process since the early 1990s, commented with respect to the authority of government and internal management in 2010 (when a New Labour government was replaced by a Conservative/Liberal Democrat coalition) and 1997 (when a Conservative government was replaced by New Labour):

“With the changes in government, we’ve been very fortunate with both Resource Accounting and Clear Line of Sight. The changes in government have not created problems. With the Clear Line of Sight, we actually spent time talking to the shadow treasury team before the introduction. We’re obviously going to seek their agreement. With Resource Accounting and Budgeting, in fact, the RAB reforms fitted perfectly with what the new government wanted to do in terms of moving to multi-year budgeting, in terms of resource DEL [department expenditure limit] control and so on.” WF1

Normalisation arguments were also presented relatively frequently within WF (NOR1 being 11.1%, Table 2). In the interviews, normalisation was often related to the role professionals play because of their specialised training and active involvement in networks, and how they influence what were viewed as appropriate processes and behaviours. There is evidence that many of the accounting (particularly financial accounting) changes that had been introduced were viewed as positive because they related to ‘what was professional’ or what ‘was done in the private sector’. In addition, WF, like the generality of Westminster interviewees, made extensive use of narratives and stories to describe and assess the changes. This evidence may have been strengthened by their long-term experience in all the three areas of accounting change, where the first experimentations were carried out already in the 1980s. This allowed the senior interviewees to expand on their answers, and provide enriched opinions, especially through the recollection of past events. Finally, it is worth noting the WF interviewees were relatively emotionally connected to the changes, with positive pathos strategies (PAT1) utilised in 7.3% of cases (Table 2). This percentage is much higher than in WD, where pathos was rarely found. The difference between the central finance function and the other departments regarding pathos may have reflected the fact that WF was, to an extent, the persuader and implementer of the accounting changes, whereas the other departments merely had the changes ‘imposed’ upon them. For example, one WF interviewee discussing the internal management and civil servants’ commitment to the overall reform process:

“I think it [RAB] moved through Treasury. I’m sure he [Andrew Likiermanvi] must have had senior support behind him at a political level as well as at civil-service level. If you had the inertia of very senior civil servants, it wouldn’t have happened. There must have been a desire above Andrew, behind Andrew that was making it happen. Everything that came to me came through Andrew. He would be very encouraging and have a very good handling on what was happening. He generated the support and the belief in everything. It must have also been coming from groups of politicians and groups of very senior civil servants. It couldn’t have happened otherwise.” WF2

14

Also at the operational level (WD), financial accounting was the frequently referenced area of change. This was even in cases where interviewees had no previous private-sector experience. When looking at the legitimation strategies in WD, the main strategy used was rationalisation (Table 2). Indeed, 30% of the arguments made were expressed on a RAT1 basis, and were especially related to the importance of economic logic, effective planning and control (especially with respect to resources), and optimal choice:

“I think it [RAB] was to demonstrate costs in and around each department because before that we weren’t; we were only thinking about the cash costs. You weren’t ever thinking about any non-cash elements before.” WD7

Interestingly, delegitimation was also often related to rationalisation (WD: RAT2, 7.5%, the highest among the delegitimation strategies, Table 2) and occurred much more frequently than in WF. Staff in WD had more doubts about the actual meaning of what they were doing and the changes they were implementing. As a consequence, changes were more often perceived as authority-based and imposed. The second most-used strategy in WD was authorisation, especially with reference to the role played by WF and government. For instance, when referring to CLOS, and on being asked about who was promoting the change:

“Treasury tell us and then we ask how high they would like us to jump.” WD1

“Change comes pretty much from the centre, which is either Treasury or the Cabinet Office. Parliament also I think. Parliament pushed CLOS through. They were fed up with getting three different reports... and every one of those is correct. And parliament, I think, just got fed up with all the smoke and mirrors and the constant trying to reconcile between the three to get facts.” WD2

Narrativisation was widely used by WD interviewees to legitimate change (Table 2). In addition, possibly reflecting the interviewees’ relatively more extensive public-sector backgrounds, the use of normalisation strategies here was much lower than in WF (NOR1 6.6%, Table 2). Finally, differently from WF, pathos arguments were almost totally absent in the WD responses (Table 2). The further the actors implementing the changes were from the centre (where changes had been decided), the less committed and personally involved they appeared to be. While decisions regarding the changes could be viewed as political, interactions between politicians and WF to shape them, and the charging of WF to implement them, resulted in a degree of emotional attachment in WF. This was missing in the operationally-focused departments.

Insert Table 2 here

Finally, the analysis of the outcome of change (as emerging from the interviewees’ perceptions) produces an interesting picture across the two organisational levels. While some of the changes were identifiable as being incremental (where modifications were made to structures and systems, but not interpretative schemes), others were perceived as more radical (involving a shift in both structures/systems and interpretive schemes). In very few cases were the accounting changes considered as resulting in no real change at all. Radical changes affecting behaviours and everyday life were often viewed as being pushed into the system through authority, but were also very frequently seen as rational and helpful. Considering the overall Westminster case (excluding

15

references to the reform process as a whole), budgeting, followed by financial accounting, changes were seen more often as radical (Table 3). Not only the formal accounting structures were introduced, but also the relative interpretive schemes changed consistently, with people agreeing with the new managerial values, recognising the usefulness of the new systems, such as accruals accounting, and using them in their day-to-day decision making. For example, with respect to RAB:

“Budgeting has been improved. I think there is more genuine information coming out. If you just use cash accounting, you are only capturing a certain amount of information. If you use accrual accounting, you are adding another dimension. I spent my career based on the assumption that good communication is important and communication has got better.” WF2

and

“Accounting was very much silos in that we had some experts, a couple of people, who knew the rules in budgeting, and they were like a breed of themselves. The accounting was done up here completely separate, and it was disjointed... Now over time with the CLOS project... it has made it more public, so that people understand that they need to be interested in how things are accounted for, and how things are treated in the budgetary framework.” WD1

Performance management, however, was often seen as positive by WD. For example, one WD interviewee argued that the move to performance management, through the embedding of PSAs and SDAs in the planning and accountability processes, and the development of multi-period planning processes, had far more impact than financial accounting changes. Interestingly, in higher education, and perhaps as a consequence of the REF, the most contentious change appeared to be relating to performance management (also referring to SDAs and PSAs). The interviewees recalled that the impetus to introduce and revise performance measures almost always came from the centre, and, as part of this process, WF often sought negotiation and consultation about details. Nevertheless, doubts about their actual implementation and meaning still remained, often leading to the perception of a change that was only formal (at the level of structures, rather than ideas and behaviours) and incremental:

“It [resource budgeting] can be affected by things outside of departmental control, the macroeconomic position; it’s a highly significant feature of managing our overall budget and managing our overall relationship with the Treasury. People are not convinced; it is contentious” WD12

The use of certain legitimation strategies, and especially seeing accounting change as a ‘rational thing to do’, appears related to the perceptions of the outcome of change. Table 4 details the co-occurrence between the perceived outcomes and the legitimation strategies in Westminster at the two organisational levels. It shows that at both levels (WF and WD), rationalisation strategies were used more often than authorisation ones, and changes were identified more frequently as radical (especially as far as financial accounting and budgeting were concerned). This suggests that legitimation through authority is necessary but not sufficient to foster radical change, as people first of all need to make sense of the change as being ‘rational’ and ‘sensible’. Consistent with this line of reasoning, in Westminster, patterns of incremental change seem to be associated with a stronger use of legitimation through authorisation (AUT1 reaching similar levels as RAT1, Table 4) and delegitimation based on rationalisation (e.g. lack of resources, skills, or culture; this being 14.1% in WF and 22.8% in WD, Table 4) and pathos (18.8% in WF and 5.3% in WD; PAT2 not present in cases of radical change).

16

Insert Tables 3 and 4 here

Legitimation of accounting changes: a comparison with Scotland

In terms of legitimation strategies, SF, perhaps unsurprisingly, differed from its immediate Westminster counterpart (WF): here, indeed, the interviewees made greater use of authorisation strategies (AUT1 representing 25.3% of the strategies used, Table 2). While the main authority mentioned by SF interviewees was usually the Scottish government (and ministers), they also often combined this with reference to the role played by WF in determining their overall resources and, often, accounting policies (such as in the case of RAB and WGA). For instance:

“The policy is devolved so ministers can set their own statutes, their own outcomes, and these sorts of things. We just have to make sure we don’t overspend and we send regular information to Treasury [WF]. We’ve built up a level of credibility with Treasury that what we say, we do. Our forecasts are pretty close to where we come out and we have all the same risk assessments too.” SF1

In SF, rationalisation and narrativisation both represented 22% of the strategies employed to describe the changes (Table 2). Interestingly, and again different from what was highlighted by WF interviewees, normalisation and pathos arguments were much less present. This could have a twofold explanation. On the one hand, although most interviewees in both Westminster and Scotland were accountants, this was relatively less the case in Scotland. In addition, fewer in Scotland had private sector experience (half in both SF and SD). This may have (negatively) impacted on their perceptions of the importance of professional standards and rules. On the other hand, with the exception of the NPF, SF did not appear to feel ownership for the changes introduced, which were usually perceived as being driven from Westminster (for example, RAB, EYF and WGA). As a consequence, lower levels of personal and organisational commitment (i.e. pathos strategies) were present. Consistently, delegitimation strategies emerged much more often here than in Westminster. Delegitimation was mainly brought forward through rational arguments (RAT2, Table 2), followed by negative stories and examples (NAR2), and critiques of the main authority figures (especially WF). For instance, criticising the rationale behind EYF and agreeing with Treasury’s decision to remove it in 2011 (joint use of NAR2, RAT2 and AUT1):

“Treasury said, ‘Well there’s end-year flexibility, you can’t just assume you can pull down what you want, when you want’. So they brought in a control and said, in that particular year, that there would be a certain fixed amount that you could access. I mean, it was still pretty generous but it was different to what people were expecting, they could just ‘willy-nilly’ access resources before. It was so relaxed; it was unbelievable!” SF1

and criticising WF and Whitehall (AUT2) in relation to the overall reform:

“We were able to have a system here where the main finance people supporting the portfolios actually are a part of the line management arrangement within my team, so you didn’t quite end up with the problems you get dealing with the Treasury and finance people in Whitehall departments.” SF2

Similar statements were not unusual in SF, where a form of “rivalry” towards the fellow accountants in Westminster (as opposed to Scotland as a devolved administration) was often identifiable.

In the Scottish operationally-focused departments (SD), narrativisation, authorisation and rationalisation were by far the most-used legitimation strategies, with narrativisation occurring

17

slightly more often than the other two (Table 2). Authorisation arguments made particular reference to the role of the EU, followed by government policies decided in both Scotland and Westminster, while rationalisation stressed efficiency and effectiveness in both the management of activities and resources. For instance, with reference to financial reporting and European regulations:

“The additional demands are effectively from Europe in terms of what they are reporting…the requirements are to make sure you satisfy all of those regulations. We have to generate duplicate activities that the core financial systems should be able to do for us but there is significant investment in terms of new management information systems to cope with the new agricultural programme.” SD2

Speaking of the reasons relating to the introduction of performance management changes (and NPF), on the basis of rationalisation arguments:

Compared to SF, SD interviewees made more pronounced use of delegitimation strategies based on rational arguments (RAT2, 10.08% vs. 9.30% in SF, Table 2), mainly with reference to financial accounting and budgeting changes. For instance, when speaking about WGA:

“I can see the, you know, the aspiration in trying to get a cohesive look at the whole of the government picture; but I don’t know what anyone could realistically use the information for.” SD1

Overall, delegitimation strategies represented about 21% of the total in Scotland, compared to 14% in Westminster (Table 2); Scottish interviewees were, thus, overall more critical of the accounting reforms put in place over the past 20 years. Finally, compared to Westminster, in Scotland change was more often identifiable as incremental, i.e. new accounting systems were implemented, but these were interpreted through the old, bureaucratic interpretive schemes and little understood (Table 3). For instance, discussing the formal implementation of RAB, and stressing that people were not actually ready and equipped for it:

“Arguably we were focused on a ‘go-live’ date and whether we were actually ready to go live on that date may be moot. We went ‘live’ arguably not really ready, there were a whole raft of issues that arose because of that.” SD5

This picture is consistent with the more frequent use of delegitimation strategies (especially through rationalisation, narrativisation, pathos and authorisation) that emerged in the devolved Scottish administration. As far as radical change in concerned, although SF interviewees indicated it more often than those in SD, at all levels this was mostly linked to the Scottish NPF (different from Westminster, where radical change was most often identifiable in respect of budgeting). For instance, in relation to the extent of change brought about by the NPF and its effect on daily decision making:

“I would genuinely say that since the introduction of the National Performance Framework, 2007, so what, about six years ago, I think during that time in government, that’s been the single biggest change that I have seen in terms of how we think about policy and how we think about delivery of policy. I think before that, it was very much about inputs and outputs.” SF5

and

“It [performance management] has changed hugely. It used to be focused on inputs. What does money buy us in terms of teachers, hospitals or hospital beds? But now we are much more focused on outcomes and

18

outputs. Outcomes ideally, but we will settle for outputs.... There has been a big shift from this input-based to more output, and you have heard of this National Performance Framework which has indicators to focus more on the outcomes. That is probably the biggest change.” SD4

There was a greater ownership felt by the interviewees for a tool (the NPF) that had been specifically crafted by the Scottish government to meet their information needs; a tool that embodied signs of the autonomy the departments have in managing and controlling their daily activities.

Table 5 details the co-occurrence between the perceived outcomes of change and the legitimation strategies in the Scottish departments and helps identify possible patterns between the use of legitimation strategies and the perceived outcome of change. Two patterns, in particular, seem to emerge also in the Scottish case. Despite the relatively lower evidence of radical change that could be found in the Scottish interviews (38% of the arguments made vs. 55% in Westminster, Table 3), radical change was, like in Westminster (but here even more clearly), mainly associated with the actors’ more frequent use of rationalisation (37.4% of the times radical change was identifiable in SF and 36.6% in SD, Table 5), rather than authorisation, strategies. Conversely, when legitimation was mainly based on authorisation arguments (rather than rationalisation ones) and occurred jointly with delegitimation, based on both pathos (9.6% in SF and 11% in SD, Table 5) and rationalisation (19.1% in SF and 24.4% in SD), incremental change was more often identifiable; this was similar to the case in Westminster.

Insert Table 5 here

Legitimating change at different government levels

As indicated, the vast majority of Westminster interviewees were professional accountants. This is fairly typical of what has been happening in the UK public sector with respect to accounting changes over the past decades, with private-sector trained professional accountants often recruited to support the implementation and operation of the new accounting systems (Connolly and Hyndman, 2006a and 2006b; Christensen and Parker, 2010). The professionalisation experience is likely to have influenced perceptions and the way in which actors’ see accounting changes. Although most of the interviewees in Scotland also had an accounting background, these were relatively fewer. This may have impacted on their perceptions and can explain some of the differences found both in the strategies used and in the perceived change at the level of interpretive schemes. Moreover, while previous literature suggested the presence of strong professional identities and rivalries during accounting change (typically between public versus private sector accountants and between accountants and other professional groups, Ellwood and Newberry, 2007; Jones, 2000; Christensen and Parker, 2010), this study does not find evidence for this. Rather, it shows that the involvement of professional accountants in the reforms of both Westminster and Scotland favoured the development of a different form of ‘rivalry’ between two different groups (central versus devolved administration) of accountants.

In Westminster, comparing the use of legitimation strategies at both organisational levels, the changes were mainly seen positively and in terms of rationalisation. This provides further evidence of the UK as being a leader and strong supporter of managerial reforms in the public sector. Consistently with the high number of accountants present, normalisation strategies and profession-based arguments were used at both organisational levels (although to a greater extent in WF). Here, the changes were first justified and their implementation throughout the whole government planned. Differently, in Scotland, there were frequent references to sources of

19

authority that were not only local (such as the Scottish government and parliament), but also UK-based (WF and the Westminster government) and supranational (like the EU). These multiple layers and levels of government continuously interacting are likely to have influenced the interviewees’ perceptions in terms of ‘who is really pushing the change’; in addition, it strengthened the role of authorisation strategies when compared to Westminster. Consistent with this interpretation and with the search for a greater local identity mirrored in the exaltation of the NPF (performance management being the area of change largely identifiable as radical), in Scotland a stronger role played by delegitimation strategies was also evident.

In terms of significance, or outcome, of change, looking across the two levels of government, consistent patterns emerge between this and the legitimation strategies used to make sense of the changes. Patterns emerge, in particular, for radical and incremental change. In both Westminster and Scotland, radical change is associated with the presence and prevalent use of arguments that link back to authorisation and rationalisation strategies to interpret the changes. Rationalisation first, and then authorisation, seem to be necessary for radical change to be identified (see Table 4 and 5). Incremental change appears, instead, characterised by both weak legitimation based on rationality and strong delegitimation, which counterbalances and more than offsets the efforts for change. The use of delegitimation strategies through the means of pathos and rationalisation arguments, in particular, hampers the change in people’s interpretive schemes, tying them to old ways of doing things. It is interesting to note that these patterns emerge consistently across organisational levels and different levels of government, regardless of the interviewees’ background (which was more accounting-related in the finance departments, much less so in the operational ones).

Finally, it is worth highlighting that legitimation through narrativisation emerged for all the outcomes of change. Each of the interviewees utilised mostly positive and constructive stories (rather than critiques and caveats) when discussing change, even in cases when the final perceived outcome may have been merely cosmetic. Perhaps surprisingly when compared to previous studies (Covaleski et al., 2003; Green, 2004; Vaara et al., 2006), in the respondents’ minds moralisation strategies seemed to be little associated with the implementation and achievement of any kind of accounting change. On the one hand, this may suggest that, in this context and contrary to their use elsewhere, accounting changes are largely seen as value neutral and rational, and not particularly related to bigger ‘moral’ questions such as appropriateness of service provision and transparency of the organisation. On the other hand, this finding could be explained by the fact that certain issues, such as ‘good governance’ and ‘transparency’, may be inbuilt in the public-sector mindset, taken for granted, and thus not in need of mention/justification in the interviewees’ minds.

Conclusions

This study aimed to improve our understanding of how new accounting techniques are implemented and translated at different organisational and government levels. It explored how accounting changes are argued for, and, by focusing on Westminster and Scotland, how this relates to the perceived outcome of change. Results show that the way organisational actors make sense of changes and legitimate them is strongly associated with the extent to which the actor’s interpretive schemes are influenced.

While previous literature suggests that different actors tend to use the same rhetorical sequences during successive periods of change, a more fine-grained analysis suggests that in the UK (Westminster and Scotland) different legitimation strategies were used at each of the organisational levels, with rationalisation and authorisation strategies dominating. The study shows that changes in accounting, originally introduced to enhance ‘rational’ management and business-like/managerial decision making, were often received as being strongly pushed by some source of ‘authority’. Both the central implementers (WF and SF), who first received the political ideas and

20

interpreted them for organisational adoption, and the more operationally-focussed departments (WD and SD), tended to provide positive (legitimating) justifications for modifications to the systems. The relatively more frequent use of pathos in WF may reflect its central position in the general UK-government landscape and the more direct connection and involvement of WF in the design of the changes. This is particularly evident when considering the greater use of negative (delegitimating) strategies made in the devolved administration; indeed, in Scotland, most of the changes (with the notable exception of NPF) were perceived as ‘centrally imposed’ by Westminster. Finally, in Westminster, and even more so in Scotland, the interviewees made substantial recourse to stories (i.e. narrativisation) to justify the changes.

Radical change (involving a shift in both structures and systems and interpretive schemes) was perceived more often in Westminster than in Scotland. This may be influenced by the different level of government considered and the different perceptions the interviewees had in terms of their autonomy in setting out and implementing change. Moreover, the line-departmental actors who did not decide the changes, but were left to engage with them and modify their everyday activities in response, tended to perceive change as much more significant and ground-breaking than those planning and shaping itfrom the centre. Overall, and regardless of the interviewees’ background and organisational level, the results suggest that radical change is more likely to take place when change is perceived, primarily, as rational and ‘sensible’, rather than merely driven by authorisation-related pressures. A certain level of authorisation, however, is important to justify the change and make it happen. In short, legitimation through authority is necessary but not sufficient to foster radical change. While authorisation is needed to implement the new systems and structures, rationalisation is essential to facilitate change at the level of interpretive schemes. Consistent with this interpretation, incremental change (where modifications are made to structures and systems, but not the interpretative schemes) is, instead, associated with relatively modest legitimation via rationalisation, and with delegitimation through the means of pathos and rationalisation.

This paper offers a threefold contribution to previous literature and theory. First, existing studies often struggle to explain organisations’ heterogeneity and practice variation and, in many cases, accounting has been assumed to converge towards the adoption of common practices (Lounsbury, 2001; Christensen and Lægreid, 2007). This study sheds more light on how individuals within organisations may legitimate accounting change and, ultimately, affect its significance within the organisation itself. It proposes a relationship between the legitimation strategies used to interpret change and its outcome, distinguishing between the level of structure and systems and interpretive schemes. How legitimation strategies may affect the reproduction and outcome of generally accepted changes has been little investigated in the past. The evidence presented here shows a relationship between strategies used and perceived significance of change, engaging in a finer-grained investigation than previously conducted. Second, the study investigates change and legitimation not only across organisational levels, but also at different government levels, looking at the Westminster government and the Scottish devolved administration. Different from previous literature (Green, 2004), the paper highlights variations in the way actors in similar organisational positions (and with similar background) interpret the same changes; these differences are mainly based on their hierarchical position (central versus devolved administration) and on the ownership they perceive of the changes being implemented (in the case of Scotland, much more felt for the NPF). The relative lack of ‘pathos’ in Scotland is likely to have influenced the limited perception of radical change. Third, the majority of previous accounting studies on accounting change have been undertaken using an organisational focus, rather than an individual one (Laughlin, 1991; Broadbent, 1992; Burns and Scapens, 2000; Liguori and Steccolini, 2012; Contrafatto and Burns, 2013). Although these studies tend to agree that radical change is more likely to occur when changes are amenable to actual and shared justification, this research highlights the active role played by organisational actors in understanding and implementing change, and explores the

21

individual legitimations that can provide such justification. It is important to highlight that individual rhetorical strategies and legitimations of change may, and indeed are likely to, be different from the official justification that the organisation provides for the introduction of a new practice.

From a managerial perspective, it is suggested that, to implement and eventually achieve radical change, those charged with implementing change need to realise that those affected, and ultimately involved in operating the new systems, have to see any adjustments as, at the very minimum, logical and sensible (rational) and as being promoted by those having the authority to do so. Moreover, radical change is more likely to occur if managers have a degree of commitment (pathos) to the change. Training and education are therefore at the core. In terms of policy implications, the study informs public-policy makers regarding linkages between new systems and already existing ones. The research shows that extant accounting approaches have long-established validity and are largely taken-for-granted; therefore it is important to consider the interaction between new and old tools when designing and introducing new accounting reforms. It is likely that any proposed abandonment, or criticism, of valued older techniques will hamper the successful implementation of new ones. In order to reduce resistance to change, the integration of new ideas into existing frameworks should be handled carefully, through appropriate, interactive, training and communication (thus impacting on people’s understanding and interpretive schemes).

As with all empirical studies, this research also has its limitations. Firstly, the study deals with organisational actors’ legitimation strategies and perceptions of change. While such perceptions are likely to be strongly correlated with the significance of change (however defined), detailed studies of systems and actions, resulting from the use of new tools and techniques, could provide further insights. Secondly, taken-for-granted issues and ideas may often remain ‘under the radar’, as not using certain words could also mean that the information is assumed and does not need to be stated (Green, 2004). Thirdly, the focus of the empirical study is on the UK, an early and strong reform initiator when compared to others. Other studies comparing different countries, possibly with differing administrative cultures and traditions, would enrich the findings. Further research could also investigate how legitimation strategies enable actors to establish and exercise different power relationships during processes of change. More studies on the relationship between outcomes and areas of change are also needed to explore how specific changes may be embedded into different organisations and cultures.

22

References

Aerts, W. (2001), “Inertia in the attributional content of annual accounting narratives”, European Accounting Review, Vol.10 No.1, pp.3-32.

Barzelay, M. (2001), The New Public Management: Improving Research and Policy Dialogue, University of California Press, Berkeley.

Bogdanor, V. (2001), Devolution in the United Kingdom, Oxford University Press: Oxford. Broadbent, J. (1992), “Change in organizations: a case study of the use of accounting information

in the NHS”, British Accounting Review, Vol.24 No.4, pp.343-367. Brown, A. D., Ainsworth, S. and Grant, D. (2012), “The rhetoric of institutional change”,

Organization Studies, Vol.33, No.3, pp.297-321. Brown, T.L. and M. Potoski (2003), “Transaction Costs and Institutional Explanations for

Government Service Production Decisions”, Journal of Public Administration Research and Theory, Vol.13 No.4, pp.441-468.

Brunsson, N. (1989), The organization of hypocrisy: Talk, decisions, and actions in organizations, John Wiley & Sons, New York.

Buhr, N. (2002), “A structuration view on the initiation of Environmental Reports”, Critical Perspectives on Accounting, Vol.13 No.1, pp.17-38.

Burns, J. and Scapens, R.W. (2000), “Conceptualising management accounting change: an institutional framework”, Management Accounting Research, Vol.11, pp.3-25.

Cho, C. (2009), ”Legitimation Strategies Used in Response to Environmental Disaster: A French Case Study of Total SA’s Erika and AZF Incidents”, European Accounting Review, Vol.18 No.1, pp.33-62.

Christensen, T. and Lægreid, P. (2007), Transcending New Public Management. The transformation of public sector reforms, Ashgate, Aldershot.

Christensen, M. and Parker, L. (2010), “Using ideas to advance professions: public sector accrual accounting”, Financial Accountability & Management, Vol.26, No.3, pp.246-266.

Clatworthy, M. and Jones, M.J. (2003), “Financial reporting of good news and bad news: Evidence from accounting narratives”, Accounting and Business Research, Vol.33 No.3, pp.171-185.

Comtois, E., Denis, J.L. and Langley, A. (2004), “Rhetorics of efficiency, fashion and politics: Hospital mergers in Quebec”, Management Learning, Vol.35 No.3, pp.303-320.

Connolly, C. and Hyndman, N. (2006a), The impact of resource accounting and budgeting in the Northern Ireland public sector, Association of Chartered Certified Accountants’ Research Report (No. 87), Certified Accountants Educational Trust, London.

Connolly, C. and Hyndman, N. (2006b), “The actual implementation of accruals accounting: Caveats from a case within the UK public sector”, Accounting, Auditing & Accountability Journal, Vol.19 No.2, pp.272-290.

Connolly, C. and Hyndman, N. (2011), “Accruals accounting in the public sector: A road not always taken”, Management Accounting Research, Vol.22 No.1, pp.26-45.

Contrafatto, M. and Burns, J (2013), “Social and environmental accounting, organisational change and management accounting: A processual view”, Management Accounting Research, Vol.24, pp.349–365.

Covaleski, M.A., Dirsmith, M.W. and Rittenberg, L. (2003), “Jurisdictional disputes over professional work: the institutionalization of the global knowledge expert”, Accounting, Organizations and Society, Vol.28, pp.323-355.

Creed, D., Scully, M. and Austin, J. (2002), “Clothes make the person? The tailoring of legitimating accounts and the social construction of identity”, Organization Science, Vol.13 No.5, pp.475-496.

Cristofoli, D., Ditillo, A., Liguori, M., Sicilia, M. and Steccolini, I. (2010), “Do environmental and task characteristics matter in the control of externalized local public services? Unveiling the relevance of party characteristics and citizens' offstage voice”, Accounting, Auditing & Accountability Journal, Vol.23, No.3, pp.350-372.

Czarniawska, B. and Joerges, B. (1996), “Travel of ideas”, in Czarniawska, B. and Sevón G. (Eds.),

23

Translating organizational change, De Gruyter, Berlin, pp.13-48. Ellwood, S. and Newberry, S. (2007),"Public sector accrual accounting: institutionalising neo-

liberal principles?", Accounting, Auditing & Accountability Journal, Vol. 20 No.4, pp. 549-573. Erkama, N. and Vaara, E. (2010), “Struggles over legitimacy in global organizational restructuring

: a rhetorical perspective on legitimation strategies and dynamics in a shutdown case.”, Organization Studies, Vol.31, No.7, pp.813-839.

Etherington, L.D. and Richardson, A.J. (1994), “Institutional pressures on university accounting education in Canada”, Contemporary Accounting Research, Vol.10 No.S1, pp.141-62.

Ezzamel, M., Hyndman, N., Johnsen, A., Lapsley, I. and Pallot, J. (2005), “Conflict and rationality: accounting in Northern Ireland’s devolved assembly“, Financial Accountability & Management, Vol.21, No.1, pp.33-55.

Fiss, P. And Zajac, E. (2006), “The Symbolic Management of Strategic Change: Sensegiving Via Framing and Decoupling”, Academy of Management Journal, Vol. 49 No. 6,pp. 1173-1193,

Green, S.E. (2004), “A rhetorical theory of diffusion”, Academy of Management Review, Vol.29 No.4, pp.653-669.

Green, S.E., Babb, M. and Alpaslan, C.M. (2008), ”Institutional field dynamics and the competition between institutional logics”, Management Communication Quarterly, Vol.22 No.1, pp.40-73.

Greenwood, R., and Hinings, C.R. (1993), “Understanding strategic change: the contribution of archetypes”, The Academy of Management Journal, Vol.36 No.5, pp.1052-1081.

Green, S.E. and Li, Y. (2011), “Rhetorical institutionalism: Language, agency, and structure in institutional theory since Alvesson 1993”, Journal of Management Studies, Vol.48 No.7, pp.1662-1697.

Greenwood, R., and Hinings, C.R. (1996), “Understanding radical organizational change: bringing together the old and the new institutionalism”, The Academy of Management Review, Vol.21 No.4, pp.1022-1054.

Hammerschmid, G. and Meyer, R. (2005), “New public management in Austria: Local Variation on a Global Theme?”, Public Administration, Vol.83 No.3, pp. 709-733.

Hinings, C.R. and Greenwood, R. (1988), The Dynamics of Strategic Change, Basil Blackwell Ltd., Oxford.

Hirsch, P.M. (1986), “From ambushes to golden parachutes: Corporate takeovers as an instance of cultural framing and institutional integration”, American Journal Sociology, Vol.91 No.4, pp.800-837.

HM Government (1982), Efficiency and effectiveness in the civil service (Cmnd. 8616), The Stationery Office, London.

HM Treasury (1994), Accounting for taxpayer’s money: Resource accounting and budgeting in government, a consultation paper (Cmnd. 2626), The Stationery Office, London.

HM Treasury (1995), Better accounting for taxpayer’s money (Cmnd. 2929), The Stationery Office, London.

HM Treasury (2011), Alexander announces new budget flexibility for devolved administrations. Available at: https://www.gov.uk/government/news/alexander-announces-new-budget-flexibility-for-devolved-administrations

Hood, C. (1995), “The ‘new public management’ in the 1980s: Variations on a theme?”, Accounting, Organizations and Society, Vol.20 No.2-3, pp.93-109.

Hood, C. and Peters, G. (2004), “The middle aging of new public management: Into the age of paradox?”, Journal of Public Administration Research and Theory, Vol.14 No.3, pp.267-282.

Hrasky, S. (2012), “Carbon footprints and legitimation strategies: symbolism or action?”, Accounting, Auditing & Accountability Journal, Vol.25 No.1, pp.174-198.

Humphrey, C. (2005), “The Questionable Nature of “Third Hand’’ Public Sector Accounting Solutions: A Case for Change?”, European Accounting Review, Vol.14, No.3, pp.475–85.

Hyndman, N. and Lapsley, I. (2016), “New Public Management: The Story Continues”, Financial Accountability & Management, Vol.32, No.4, pp. 385-408.

24

Hyndman, N., Liguori, M., Meyer, R., Polzer, T., Rota, S. and Seiwald, J. (2014), “The translation and sedimentation of accounting reforms. A comparison of the UK, Austrian and Italian experiences”, Critical Perspective on Accounting, Vol.25, No.4-5, pp.388-408.

Jones, R. (2000), “National accounting, government budgeting and the accounting discipline”, Financial Accountability&Management, Vol.16 No.2, pp. 101-116.

Joutsenvirta, M. and Vaara, E. (2015), “Legitimacy Struggles and Political Corporate Social Responsibility in International Settings: A Comparative Discursive Analysis of a Contested Investment in Latin America”, Organization Studies, Vol. 36, pp.741-777.

Kitchener, M. (1999), “All fur coats and no knickers: Contemporary organizational change in United Kingdom hospitals”, in Brock, D. (Ed.), Restructuring the professional organization: Accounting, health care and law, Routledge, London.

Knights, D., and Collinson, D.L. (1987), “Disciplining the shopfloor: a comparison of the disciplinary effects of managerial and financial accounting”, Accounting, Organizations and Society, Vol.12 No.5, pp.457-477.

Koopmans, R. and Statham, P. (1999), Ethnic and Civic Conceptions of Nationhood and the Differential Success of the Extreme Right in Germany and Italy. How Social Movements Matter, University of Minnesota Press, Minneapolis.

Lapsley, I. (2009), “New public management: The cruellest invention of the human spirit?”, Abacus, Vol.45 No.1, pp.1-21.

Lapsley, I. and Midwinter, A. (2008), “Accounting and Democratic Accountability in Scotland”, in Ezzamel, M., Hyndman, N., Johnsen, A. and Lapsley, I. (Eds.), Accounting in Politics: Devolution and democratic accountability, Routledge, London.

Laughlin, R. (1991), “Environmental disturbances and organizational transitions and transformations: some alternative models”, Organization Studies, Vol.12 No.2, pp.209-232.

Lefsrud, L.M. and Meyer, R.E. (2012), “Science or science fiction? Professionals’ discursive construction of climate change”, Organization Studies, Vol.33 No.11, pp.1477-1506.

Liguori, M. (2012a), “The supremacy of the sequence: Key elements and dimensions in the process of change”, Organization Studies, Vol.33 No.4, pp.507-39.

Liguori, M. (2012b), “Radical change, accounting and public sector reforms: a comparison of Italian and Canadian municipalities”, Financial Accountability & Management, Vol.28 No.4, pp.437-463.

Liguori, M. and Steccolini, I. (2012), “Accounting change: Explaining the outcomes, interpreting the process”, Accounting, Auditing & Accountability Journal, Vol. 25 No.1, pp.27-70.

Lounsbury, M. (2001), “Institutional sources of practice variation: Staffing college and university recycling programs”, Administrative Science Quarterly, Vol.46, pp.29-56.

McNulty, T. and Ferlie, E. (2004), “Process transformation: Limitations to radical organizational change within public service organizations”, Organization Studies, Vol.25 No.8, pp.1389-1412.

Mobus, J. (2005), “Mandatory environmental disclosures in a legitimacy theory context”, Accounting, Auditing & Accountability Journal, Vol.18 No.4, pp.492 – 517.

Motion, J. and Leitch, S. (2009), “The transformational Potential of Public Policy Discourse”, Organization Studies, Vol.31, pp.1-17.

Neu, D., Cooper, D.J. and Everett, J. (2001), “Critical Accounting Interventions”, Critical Perspectives on Accounting, Vol.12 No.6, pp.735-762.

Patton, M.Q. (2002), Qualitative Research and Evaluation Methods, Sage Publications, London. Pettigrew, A. (1990), “Longitudinal field research on change: theory and practice”, Organizational

Science, Vol.1 No.3, pp.267-92. Phillips, N., Lawrence, T. Hardy, C. (2004), “Discourse and Institutions”, Academy of Management

Review, Vol. 29 No. 4, pp. 635-652. Pollitt, C. and Bouckaert, G. (2011), Public sector reform. A comparative analysis: New public management,

governance, and the neo-Weberian state (3rd ed.), Oxford University Press, Oxford.

25

Poole, M., Scott, A., Van de Ven, K. and Holmes, M.E. (2000), Organizational change and innovation processes: Theory and methods for research, Oxford University Press, New York.

Ragin, C. (2000), Fuzzy-set Social Science, The University of Chicago Press, Chicago. Sahlin-Andersson, K. (1996), “Imitating by editing success. The construction of organizational

fields and identities”, in Czarniawska, B. and Sevón, G. (Eds.), Translating organizational change. De Gruyter, Berlin, pp.69–92.

Scott, W.R. (1995), Institutions and organizations, Sage, Thousand Oaks, CA. Scottish Government (2011), An Introduction to Scotland's National Performance Framework, APS Group

Scotland, Scotland, available at: http://www.scotland.gov.uk/Resource/Doc/933/0124200.pdf Scottish Office (1998), Report on the Financial Issues Advisory Group, Stationery Office, Edinburgh. Scottish Parliament (2016), Devolved and Reserved Matters, available at:

http://www.scottish.parliament.uk/visitandlearn/Education/18642.aspx Siti-Nabiha, A.K. and Scapens, R.W. (2005), “Stability and change: an institutionalist study of management accounting change”, Accounting, Auditing & Accountability Journal, Vol.18 No.1, pp.44-73. Strang, D. and Meyer, J.W. (1994), “Institutional conditions for diffusion”, in Scott, R.W. and

Meyer, J.W. (Eds.), Institutional environments and organizations: Structural complexity and individualism, Sage, Thousand Oaks, CA.

Suchman, M.C. (1995), “Managing legitimacy: Strategic and institutional approaches”, Academy of Management Review, Vol.20 No.3, pp.729-757.

Suddaby, R. and Greenwood, R. (2005), “Rhetorical strategies of legitimacy”, Administrative Science Quarterly, Vol.50 No.1, pp.35-67.

ter Bogt, H.J. and Scapens, R.W. (2014), “Institutions, Rationality and Agency in Management Accounting: Rethinking and Extending the Burns and Scapens Framework”, research paper available at SSRN, http://dx.doi.org/10.2139/ssrn.2464980. Tsang, E.W.K. (2002), “Self-serving attributions in corporate annual reports: A replicated study”, Journal of Management Studies, Vol.39 No.1, pp.51-65.

Vaara, E. (2002), “On the discursive construction of success/failure in narratives of postmerger integration”, Organization Studies, Vol.23 No.2, pp.213-250.

Vaara, E., Tienari, J. and Laurila, J. (2006), “Pulp and paper fiction: On the discursive legitimation of global industrial restructuring”, Organization Studies, Vol.27 No.6, pp.789-810.

Van Leeuwen, T. and Wodak, R. (1999), “Legitimizing Immigration Control: A discourse-historical analysis”, Discourse Studies, Vol.1 No.1, pp.83-118.

26

Tables

Table 1 – Main accounting changes in the UK central government systems 1980s 1990s 2000s

Performance management

1982 - Focus on performance management in Financial Management Initiative. 1988 - Intensification of performance focus with the introduction of the Next Steps Initiative and the creation of Executive Agencies.

1994-1995 - Announced that Statements of Resources by Departmental Aims and Objectives to be introduced in all government departments as part of the Resource Accounting and Budgeting (RAB) changes. 1998 - Performance-focused Public Service Agreements and Service Delivery Agreements (SDAs) introduced as a key-aspect of quasi-contracts between the Treasury and government departments.

2001 - Statements of Resources by Departmental Aims and Objectives required to be produced as part of the RAB changes in all central government departments. 2004 - SDAs replaced by Delivery Plans for each government department.

Financial Accounting

1980 - Largely cash based and mainly focusing on cash received and cash spent.

1994-1995 - Announced that resource (accruals) accounting to be introduced in all government departments as part of the RAB changes. 1998 - Treasury first published a scoping study for Whole of Government Accounts (WGA).

2001 - Resource (accruals) accounting ‘live’ in all central government departments. 2007 - Announced that financial accounting of government departments to be based on International Financial Reporting Standards (IFRS). 2009 - Financial accounting of government departments on basis of IFRS. 2011 - First set of WGA.

Budgeting 1980 - Based on cash and principle of annuality.

1994-1995 - Announced that resource (accruals) budgeting to be introduced in all government departments as part of the RAB changes. 1997 - End-year flexibility (EYF) introduced with respect to budget carry forward. 1998 - Treasury-led spending reviews initiated to set firm and fixed spending budgets over several years for each government department.

2003 - Resource (accruals) budgeting ‘live’ in all central government departments. 2007 – ‘Clear Line of Sight’ (CLOS) project commenced aimed at getting better alignment of accounting and budgeting information. 2011 - EYF system abolished due to financial pressures (replaced by a system of Budget Exchange).

27

Table 2 – Legitimation (and delegitimation) strategies in Westminster and Scotland

WESTMINSTER

AUT1 AUT2 MOR1 MOR2 NAR1 NAR2 NOR1 NOR2 PAT1 PAT2 RAT1 RAT2 Total counts

WF

22.2% 1.9% 0.3% 0.00% 19.7% 2.9% 11.1% 0.3% 7.3% 4.1% 24.4% 5.7% 100%

70 6 1 0 62 9 35 1 23 13 77 18 315

WD

26.6% 1.6% 3.8% 0.00% 18.8% 2.1% 6.6% 0.4% 0.6% 2.1% 30.0% 7.5% 100%

205 12 29 0 145 16 51 3 5 16 231 58 771

Total counts

25.3% 1.7% 2.8% 0.00% 19.1% 2.3% 7.9% 0.4% 2.6% 2.7% 28.4% 7.0% 100%

275 18 30 0 207 25 86 4 28 29 308 76 1086

SCOTLAND

AUT1 AUT2 MOR1 MOR2 NAR1 NAR2 NOR1 NOR2 PAT1 PAT2 RAT1 RAT2 Total counts

SF

25.3% 3.4% 2.0% 0.3% 22.0% 3.9% 6.0% 0.4% 2.3% 2.9% 22.0% 9.3% 100%

173 23 14 2 151 27 41 3 16 20 151 64 685

SD

22.4% 2.7% 2.9% 0.0% 23.3% 4.2% 6.3% 0.9% 0.9% 3.5% 22.9% 10.1% 100%

149 18 19 0 155 28 42 6 6 23 152 67 665

Total counts

23.9% 3.0% 2.4% 0.1% 22.7% 4.1% 6.1% 0.7% 1.6% 3.2% 22.4% 9.7% 100%

322 41 33 2 306 55 83 9 22 43 303 131 1350

28

Table 3 – Accounting change: perceived outcome by area of change in Westminster and Scotland

Change outcome → Radical Incremental No change

Counts

%

Area of accounting change ↓ Westminster

FIN ACC 52.5% 43.4% 4.0% 100%

52 43 4 99

BDGT 58.1% 34.4% 7.5% 100%

54 32 7 93

PM 48.6% 45.7% 5.7% 100%

17 16 2 35

Whole reform 64.0% 36.0% 0.0% 100%

16 9 0 25

Westminster 55% 40% 5% 100%

Overall count 139 100 13 252

Area of accounting change ↓ Scotland

FIN ACC 33.0% 65.0% 2.0% 100%

33 65 2 100

BDGT 31.9% 64.6% 3.5% 100%

36 73 4 113

PM 54.9% 45.1% 0.0% 100%

28 23 0 51

Whole reform 57.1% 42.9% 0.0% 100%

12 9 0 21

Scotland 38% 60% 2% 100%

Overall count 109 170 6 285

29

Table 4 – Co-occurrences of perceived accounting change outcomes and legitimation strategies in Westminster

Strategy →

AUT1 AUT2 MOR1 NAR1 NAR2 NOR1 NOR2 PAT1 PAT2 RAT1 RAT2

Counts

Change outcome ↓

%

WF (n)

Radical

13.0% 0.0% 0.0% 26.1% 0.0% 4.3% 0.0% 2.2% 0.0% 52.2% 2.2% 100%

6 0 0 12 0 2 0 1 0 24 1 46

Incremental

14.1% 3.1% 0.0% 15.6% 7.8% 6.3% 0.0% 4.7% 18.8% 15.6% 14.1% 100%

9 2 0 10 5 4 0 3 12 10 9 64

No change

33.3% 0.0% 0.0% 16.7% 16.7% 0.0% 0.0% 16.7% 0.0% 16.7% 0.0% 100%

2 0 0 1 1 0 0 1 0 1 0 6

Overall count 17 2 - 23 6 6

- 5 12 35 10 116

WF 14.7% 1.7% 0.0% 19.8% 5.2% 5.2% 0.0% 4.3% 10.3% 30.2% 8.6% 100%

Change outcome↓ WD

Radical

13.8% 0.0% 4.6% 26.2% 0.0% 3.8% 0.0% 2.3% 0.0% 46.9% 2.3% 100%

18 0 6 34 0 5 0 3 0 61 3 130

Incremental

15.8% 1.8% 3.5% 19.3% 8.8% 3.5% 0.0% 0.0% 5.3% 19.3% 22.8% 100%

9 1 2 11 5 2 0 0 3 11 13 57

No change

0.0% 0.0% 0.0% 16.7% 0.0% 0.0% 0.0% 0.0% 16.7% 0.0% 66.7% 100%

0 0 0 1 0 0 0 0 1 0 4 6

Overall count 27 1 8 46 5 7 0 3 4 72 20 193

WD 14% 0.5% 4.1% 23.8% 2.6% 3.6% 0.0% 1.6% 2.1% 37.3% 10.4% 100%

30

Table 5 – Co-occurrences of perceived accounting change outcomes and legitimation strategies in Scotland

Strategy →

AUT1 AUT2 MOR1 MOR2 NAR1 NAR2 NOR1 NOR

2 PAT1 PAT2 RAT1 RAT2

Counts

Change outcome ↓

%

SF (n)

Radical

17.8% 0.9% 0.9% 0.0% 31.8% 0.0% 3.7% 0.0% 6.5% 0.0% 37.4% 0.9% 100%

19 1 1 0 34 0 4 0 7 0 40 1 107

Incremental

19.1% 5.1% 1.5% 0.7% 12.5% 11.0% 3.7% 0.7% 0.7% 9.6% 16.2% 19.1% 100%

26 7 2 1 17 15 5 1 1 13 22 26 136

No change

50.0% 50.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 100%

1 1 0 0 0 0 0 0 0 0 0 0 2

SF 18.8% 3.7% 1.2% 0.4% 20.8% 6.1% 3.7% 0.4% 3.3% 5.3% 25.3% 11.0% 100%

Overall count 46 9 3 1 51 15 9 1 8 13 62 27 245

Change outcome↓ SD

Radical

18.3% 0.0% 4.3% 0.0% 24.7% 0.0% 4.3% 1.1% 5.4% 0.0% 36.6% 5.4% 100%

17 0 4 0 23 0 4 1 5 0 34 5 93

Incremental

15.7% 0.8% 1.6% 0.0% 11.8% 16.5% 6.3% 0.0% 0.0% 11.0% 11.8% 24.4% 100%

20 1 2 0 15 21 8 0 0 14 15 31 127

No change

16.7% 0.0% 0.0% 0.0% 33.3% 0.0% 0.0% 0.0% 0.0% 16.7% 16.7% 16.7% 100%

1 0 0 0 2 0 0 0 0 1 1 1 6

SD 16.8% 0.4% 2.7% 0.0% 17.7% 9.3% 5.3% 0.4% 2.2% 6.6% 22.1% 16.4% 100%

Overall count 38 1 6 0 40 21 12 1 5 15 50 37 226

31

Appendix

Coding scheme and examples of quotations

Outcome of change

Examples of quotations/area of change (identifying interviewee)

Radical “They embrace that [RAB] now in the knowledge that it comes with implications. They ask the right questions about this and those factors will be taken into consideration in how they are going to deploy it.” (WF2: Accounting, Budgeting)

“Obviously there was a lack of understanding of what it [RAB] meant and I still think, it still persists. I remember I used to work on introducing a new budget database that was going to manage a lot of our reporting to Treasury and we used to go up and visit a guy who worked in Rural Affairs. He was always harking on about the elegant simplicity of the old ways and was really pretty resistant to the whole thing.” (SD5: Accounting, Budgeting)

Incremental

Strategy (1- positive, 2-

negative) Examples of quotations/coding (identifying interviewee)

Authorisation “That has had a very dramatic effect on departments and it’s something that permanent secretaries have been much more animated about because it has made it difficult for them to spend money in ways that they otherwise might have spent money because they’ve now got two or three extra loops of approvals to go through. And if they want to go and recruit somebody from, wherever, they need to go and seek the state’s approval and then Francis Maude’s [a minister] approval, for one interim to work for ten months. So that’s a kind of castration of authority that’s never been there before for the civil service.” (WD2: AUT2)

Rationalisation/logos

“What’s really difficult is measuring the real outputs with your money. You know it might be how many children go to school, you can tell that, but if you say how many businesses did you help, or how many did you take in from bankruptcy or whatever, you never really see those numbers. We’ve got so many elements in our businesses that are very difficult to measure. So I think we’re still sort of seeking ways to better align what are you getting for your money in terms of actual output.” (SD6: RAT1).

Normalisation/ethos

“There is more of it [change] going on in some places than others, but unfortunately in some corners there’s not much of it happening at all and again, it’s why industry is so much better at it – it has to satisfy a bunch of shareholders who demand improvement, growth and constantly greater efficiency and all those things.” (WF1: NOR1)

Pathos “There have been several benefits and the quality of engagement that we are having with the institution has increased. We are now talking about collective contribution to national aspiration. The sector itself sees it as an opportunity to demonstrate to government what they are delivering for a significant amount of public funds which they see as putting them in a better place for spending review time. They are not pulling out research or economists’ reports out of a drawer and telling them here is how wonderful we are… they are actually drawing on something which is part of the day to day process...this is what you asked us to deliver and this is us demonstrating that we are delivering”. (SD6: PAT1)

Moralisation “You know, when most people think about rural development it’s about people coming together. Whether it’s to preserve a village hall and much more; but at the same time we give a third of the budget, so a third of £1.2billion, went to farmers for maintaining land. We are required to review less favoured areas…” (SD4: MOR1)

Narrativisation “How useful people looking at them [performance measures] find them? I suppose what they inevitably are, at least, is an indicator of whether something is going in the right direction or is going in the other direction. You know, external, visible indicators. So for example, I suppose the employment picture, the economic-type picture, and some of that. The Scottish government doesn’t produce a nice glossy annual report that says “this is what we said we’d do; and this is what we’ve done” […] Does absolutely everything work so that everyone can follow it though? No, and that would be lovely.” (SF3: NAR2)

32

i It is important, although beyond the scope of this paper, to note that legitimation strategies do not only represent rhetorical devices available to the actors to translate and understand change, but can also be used by individuals to gain or exercise power within an organisation. We thank one of the reviewers for pointing this out. ii It is important to stress that this paper adopts the definition of outcome of change (incremental vs. radical) as the movement between archetypes. We follow the wording and definitions provided by previous archetype theory literature; it is not our aim to assess the success of change. Sometimes the word ‘outcome’ can be used as an element within a management (or performance) system consisting of inputs, activities, outputs and outcomes, with outcome defined in terms the long-term effect of a change on an organisation’s effectiveness and efficiency. This is not its meaning in the context of this research. iii It is interesting to note that Laughlin (1991) and Broadbent (1992), although using a different theoretical lens (Habermasian concepts), identify and explore ‘first order’ (akin to incremental change in this study) and ‘second order’ (akin to radical change) change concepts. A major difference in the approach of these authors is that they consider the organisational level rather than the individual level (which is the focus in this study). iv In order to avoid unnecessary length, only summary information about the interviewees and examples of the analysed documents is provided in the paper. More information is available from the authors on request. v In some cases, while coding, it was not possible to identify any outcome of change from the interviewees’ statements. In these instances, a residual category was coded as ‘outcome not mentioned’. For presentational purposes, this has not been reported in the paper. vi Andrew Likierman was Head of the UK Government Accountancy Service from 1993 until 2004. He was responsible for the development and implementation of RAB.